CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Multiphase modelling using Fluent VOF: Solver settings (

akash February 22, 2013 07:29

Multiphase modelling using Fluent VOF: Solver settings

I'm trying to simulate shape oscillations of a drop in fluid. The two fluids I'm using are defined using user-defined properties to get required Reynolds number and Webers number. Its is a transient 2D problem where the shape of the drop, initially perturbed changes w.r.t. time to attain circular shape due to surface tension. But due to the inertia it starts oscillating and the amplitude decays due to the presence of viscosity.

I need to monitor the frequency and the amplitude of the oscillations. And I'm not getting the required result. The results are severely off from the analytical solution. I guess I'm setting some solver parameter wrongly. Please help me identify and also any suggestions are welcome.

I'm mentioning all the settings I'm using, step by step. (I'm using Ansys 14)

1) Import the grid, scale it to 'cm'. Set units to 'cgs'. Adapt a region (circular-0.75cm).
2)Set 'Pressure based' and 'Transient' solver. (Pressure based because both the fluids are incompressible)
3)Select 'VOF model' in multiphase model and use the default settings. (Explicit VOF, because I need time accurate solution)
4)Viscous model is 'Laminar'.
5)In materials I copy my user defined materials.
6)Set the primary and secondary phase, and also set the surface tension in the 'interaction' option.
7)Set the boundary and operating conditions.
8)Solution methods - Pressure-velocity coupling: PISO
Spatial Descritization- Gradient:Least Squares cell based
Pressure: PRESTO!
Momentum: Second order upwind
Volume Fraction: Geo-reconstruct
Transient Formulation- First order Implicit
9)Initialize solution from all zones.
10)Fixed time step. Time step of 0.001. Run simulation for 10000 time steps. Max iter/step-50.
The things which I have not mentioned, it means I have used the default settings.

Actually the solution converges at each time step. Also the solution is quite physically feasible. But the frequency of oscillation is not matching at all.

Can anyone let me know where I'm going wrong. :(

All times are GMT -4. The time now is 09:50.